Improved reservoir characterisation using fuzzy logic platform: an integrated petrophysical, seismic structural and poststack inversion study Muhammad Kamran Jafri 1,2 Aref Lashin 3,4,6 El-Khedr Hassan Ibrahim 1,5 Kamal A. Hassanein 1 Nassir Al Arifi 1 Muhammad Naeem 1 1 King Saud University, College of Science, Department of Geology and Geophysics, PO Box 2455, Riyadh 11451, Saudi Arabia. 2 Bahria University, Department of Earth and Environmental Sciences, Islamabad 44000, Pakistan. 3 King Saud University, College of Engineering - Petroleum and Gas Engineering Department, PO Box 800, Riyadh 11421, Saudi Arabia. 4 Benha University, Faculty of Science - Geology Department, Benha 13518, Egypt. 5 Mansoura University, Faculty of Science - Geology Department, Mansoura 35516, Egypt. 6 Corresponding author. Email: aref70@hotmail.com Abstract. There is a tendency for applying different integrated geophysical approaches for better hydrocarbon reservoir characterisation and interpretation. In this study, petrophysical properties, seismic structural and poststack seismic inversion results are integrated using the fuzzy logic AND operator to characterise the Tensleep Sandstone Formation (TSF) at Powder River Basin (PRB), Wyoming, USA. TSF is deposited in a coastal plain setting during the Pennsylvanian era, and contains cross-bedded sandstone of Aeolian origin as a major lithology with alternative sabkha dolomite/ carbonates. Wireline logging datasets from 17 wells are used for the detailed petrophysical evaluation. Three units of the TSF (A-sandstone, B-dolomite and B-sandstone) are targeted and their major rock properties estimated (i.e. shale/clay volume, V sh ; porosity, f Eff ; permeability, K; fluid saturations, S w and S H ; and bulk volume water, BVW). The B-sandstone zone, with its petrophysical properties of 5–20% effective porosity, 0.10–250 mD permeability and hydrocarbon potential up to 72%, is considered the best reservoir zone among the three studied units. Distributions of the most important petrophysical parameters of the B-sandstone reservoir (V sh , f Eff , K, S w ) are generated as GIS thematic layers. The two-dimensional (2D) and three-dimensional (3D) seismic structural interpretations revealed that the hydrocarbons are entrapped in an anticlinal structure bounded with fault closures at the west of the study area. Poststack acoustic impedance (PSAI) inversion is performed on 3D seismic data to extract the inverted acoustic impedance (AI) cube. Two attribute slices (inverted AI and seismic amplitude) were extracted at the top of the B-sandstone unit as GIS thematic layers. The reservoir properties and inverted seismic attributes were then integrated using fuzzy AND operator. Finally, a fuzzy reservoir quality map was produced, and a prospective reservoir area with best reservoir characteristics is proposed for future exploration. The current study showed that integration of petrophysical, seismic structural and poststack inversion under a fuzzy logic platform can be used as an effective tool for interpreting multiple reservoir zones. Key words: fuzzy logic, petrophysical analysis, reservoir characterisation, seismic inversion, Teapot Dome, Tensleep. Received 27 October 2015, accepted 29 April 2016, published online฀23฀June฀2016 Introduction Effective description of petroleum reservoirs holds great importance in hydrocarbon exploration as well as reservoir management and development plans. Reservoir characterisation is an integrated approach to describe the reservoir as subunits based on its rock properties (Amaefule et al., 1993). Well logs, cores, production or engineering data and seismic data can be integrated and used as an input to describe the reservoir characteristics. At any stage of hydrocarbon exploration to exploitation, well logs and three-dimensional (3D) seismic data can be integrated to efficiently characterise the reservoir for achieving economic benefits (Amigun and Bakare, 2013). Core data are of great importance and give direct measurement of rock properties compared to the wireline logs (Sneider and King, 1984; Amaefule et al., 1993). Seismic data can be used in delineating the prevailing structural elements that affect the reservoir of interest and determining its exact subsurface geometry. The velocity of seismic waves depends mainly on the elastic and petrophysical properties (pore spaces, matrix type, fluid content, lithology content and type, etc.) of the studied rock units. The traveltimes (one- and two-way) and the average and interval velocities are very important parameters in velocity analysis due to their role in clarifying the boundaries of the different rock units and defining the acoustic subsurface characteristics of the study area, with special emphasis on the hydrocarbon-bearing sections (Gardner et al., 1974; Wang, 2001; Lashin and Abd El Aal 2004a, 2004b; Contreras et al., 2005; El-Mowafy and Marfurt, 2008; El-Naby et al., 2009; Lashin et al., 2011, 2014; AlMuhaidib et al., 2012; Lashin and El-Naby, 2014). CSIRO PUBLISHING Exploration Geophysics http://dx.doi.org/10.1071/EG15112 Journal compilation Ó ASEG 2016 www.publish.csiro.au/journals/eg